Background: Central line-associated bloodstream infections (CLABSI) and surgical site infections (SSI) as defined by and reported to CDC's National Healthcare Safety Network (NHSN) have been added to the Centers for Medicare and Medicaid Services' new pay-for-reporting requirements. These healthcare-associated infections (HAIs) are largely preventable but may represent a small fraction of all HAIs. CDC's HAI and antimicrobial use prevalence survey will complement NHSN data and fill knowledge gaps by addressing all HAI types and antimicrobial use throughout acute-care hospitals. In 2010, a preliminary survey was conducted with CDC's Emerging Infections Program (EIP).
Objective: To determine HAI and antimicrobial use prevalence in a sample of acute-care inpatients in 10 EIP sites (CA, CO, CT, GA, MD, MN, NM, NY, OR, TN) and refine methods for a 2011 full-scale survey.
Methods: One to 3 hospitals in each site voluntarily participated. Hospital personnel performed the survey on 1 day, collecting demographic, device and antimicrobial use data on a random sample of patients. EIP staff reviewed medical records of patients identified by hospital personnel as being on antimicrobials to collect antimicrobial data and assess for active HAIs using NHSN definitions. Data collection and entry into a web-based data management system was completed in 12/2010. Descriptive statistics were calculated using OpenEpi 2.3.1 and SAS 9.2.
Results: 2015 patients in 22 hospitals were surveyed; there were 136 HAIs (Table) for a prevalence of 6.8% (95% CI: 5.7–7.9%). Pneumonia (PNEU, 32.3%) and SSI (20.6%) were the most prevalent HAIs; only 6.6% were CLABSIs. Staphylococcus aureus (SA) was the most common pathogen (29/136, 21.3%); 44.8% of SA isolates were methicillin-resistant. Antimicrobial use prevalence was 48.3% (95% CI: 46.2–50.5%). In 731 patients receiving treatment for active infection, vancomycin (218, 29.8%) and piperacillin/tazobactam (139, 19.0%) were the most commonly administered antimicrobials.
Conclusions: HAI prevalence in this survey was similar to a 2009 single-city pilot survey as well as previously published CDC analyses, and broad-spectrum antibiotic use was common. Our data suggest that initiatives focusing on CLABSI and SSI surveillance and reporting may address only about a quarter of HAIs in U.S. hospitals. Prevalence surveys may provide information useful for targeting surveillance and prevention efforts. A full-scale survey in a large sample of U.S. acute care inpatients is planned for 2011.
Table: HAI distribution (N=136).
HAI | N (%) | Device associated, N (%) |
PNEU | 44 (32.3) | 21 (47.7) |
SSI | 28 (20.6) | - |
Gastrointestinal infection* | 17 (12.5) | - |
BSI | 16 (11.8) | 9 (56.3) |
Urinary tract infection | 16 (11.8) | 12 (75.0) |
Lower respiratory infection | 5 (3.7) | - |
Skin/soft tissue infection | 4 (2.9) | - |
Eye/ear/nose/throat infection | 3 (2.2) | - |
Other | 3 (2.2) | - |
*Includes Clostridium difficile